Refractory epilepsy is a major health problem: roughly one-third of epilepsy patients (or approximately one in 300 people) have seizures that do not respond to standard medical treatments. Nutritional therapy, specifically a low carbohydrate ketogenic diet, can be an effective treatment for these patients [unreadable] indicating a strong link between metabolism and seizure susceptibility [unreadable] but the strict diet is quite difficult, and the mechanism(s) of efficacy in animal models have remained elusive. This project will explore the link between metabolism and seizure susceptibility through an alternative approach that does not require dietary manipulation. We have found that mutations of a single gene, Bad, can produce striking resistance to seizures in mice. The BAD protein is best known for its role in regulating the mitochondrial pathways of apoptosis, but the pattern of mutational effects indicates that seizure resistance is related instead to the newly identified non-apoptotic role of BAD in metabolism. BAD has been found to regulate the cellular choice of fuel consumption through a phosphorylation dependent mechanism that enables metabolism of glucose as compared with other carbon substrates such as ketone bodies and fatty acids. The studies proposed here test the hypothesis that BAD reprograms energy metabolism to determine the choice of fuel substrates in the brain, which may serve as a trigger for changes in neuronal excitability and seizure susceptibility. We will employ a multi-disciplinary approach to test this hypothesis at three levels. In aim 1, we will examine BAD-dependent alterations in cellular energy metabolism in neurons and glial cells, using strategies that include real time assessment of mitochondrial fuel oxidation, live cell imaging using novel metabolic biosensors, and mass spectrometry based profiling of metabolite byproducts of fuel consumption. These studies will dissect the mechanistic underpinnings of BAD modulation of neurometabolism that may influence excitability. In aim 2, we will perform electrophysiological studies in brain slices to directly measure the consequences of BAD manipulation for cellular and network excitability in brain slices, guided by our preliminary findings that suggest increased activity of metabolically sensitive KATP channels in neurons of Bad mutant mice. In aim 3, we will test the ability of Bad mutations to influence seizures in vivo, using continuous video-EEG monitoring, in several acute seizure models as well as in a mouse genetic model of chronic epilepsy with spontaneous seizures. Taken together, these aims will allow vertical integration in understanding BAD[unreadable]s effect on neuroenergetics and seizure susceptibility. In addition to genetic tools for manipulating metabolism by mutations in the Bad gene, novel pharmacomimetic compounds modeled after the metabolically active domain of the BAD protein will be used at each stage of the studies. These compounds may offer a future path for translating our improved understanding of the link between metabolism and excitability into potential new therapies for epilepsy.